Most of the four million newborn infants who die each year are preterm or low birth weight neonates in low‐ and middle‐income countries in south‐central Asia and sub‐Saharan Africa (Lawn 2005; Zupan 2005). Acquired infection is one of the major preventable causes of global neonatal mortality. Invasive infection in newborn infants also causes serious morbidity, including adverse effects on growth and neurodevelopment, and consumes scarce health care resources (Stoll 1997; Zaidi 2005). There is a need for inexpensive community‐ and health care facility‐based strategies to prevent serious infection in vulnerable newborn infants in low‐ and middle‐income countries (Costello 2004; Osrin 2004; Darmstadt 2005).

One such potential intervention is topical emollient therapy to improve skin barrier function (Darmstadt 2002a). The immature skin of preterm infants is ineffective as an epidermal barrier and is easily traumatised so providing a route for invasion by skin bacteria and fungi (Rutter 1988; Evans 1986). Furthermore, preterm infants lack vernix caseosa which contains anti‐microbial polypeptides that protect against invasion by micro‐organisms (Yoshio 2003). As well as providing a physical barrier to skin disruption, emollient oils, creams, or ointments provide lipids that are integrated into the epidermis to further enhance skin barrier function. Topical oils may also be a nutritional source of essential fatty acids for preterm or low birth weight infants (Lee 1993; Solanki 2005).

A Cochrane review of randomised controlled trials undertaken in high‐income countries found that prophylactic application of topical emollient increased the risk of nosocomial infection in preterm infants (Conner 2003). Most participating infants were of extremely low birth weight and most infections were due to coagulase‐negative staphylococci. The emollient most commonly used in the trials was aquaphor ointment. It is possible that this provided an environment conducive to the proliferation of micro‐organisms on the immature epidermis.

The epidemiology of invasive infection in preterm or low birth weight infants in low‐ or middle‐income countries differs in several key respects from that in high‐income countries:
1. Most infections are due to Gram‐negative bacteria and the attributable mortality and morbidity is much greater than that of coagulase‐negative staphylococcal infection (Stoll 1997; Stoll 2003; Zaidi 2005).
2. The population of infected infants are generally more mature than the extremely preterm and extremely low birth weight infants who participated in the trials in high‐income countries (Conner 2003).
3. In low and middle‐income countries, acquired infection is much less likely to be associated with intensive care or invasive procedures (Zaidi 2005).
4. Many preterm or low birth weight infants are cared for in a community or home environment as well as in health‐care facilities.
5. The infection risk may be even greater in communities in those low and middle‐income countries where maternal human immunodeficiency virus (HIV) infection is common since maternal HIV infection is associated with adverse perinatal outcomes including preterm delivery, low birth weight, and intrauterine growth restriction (Brocklehurst 1998).

Applying emollients, usually readily available and inexpensive natural vegetable or plant oil, to the newborn infant's whole body surface is a widespread traditional neonatal practice in many low‐ and middle‐income countries. Topical emollient therapy is perceived to have several potential benefits including prevention of infection and hypothermia (Darmstadt 2002b; Mullany 2005). However, although emollients may plausibly improve skin‐barrier function, there is potential for the process of application, usually by massage, to adversely affect skin integrity so increasing the risk of invasive infection. Mustard oil is the most commonly used emollient, particularly in south Asia, but other natural plant oils (for example, safflower, sesame, coconut, olive, and soybean oils) are also commonly available and used. Studies using animal models have raised concern that mustard oil is much less effective at promoting and maintaining skin integrity than other emollients, particularly sunflower seed oil (Darmstadt 2002a).

Given the potential for topical emollient therapy to prevent infection in preterm or low birth weight infants in low‐ or middle‐income countries, but also the risk that this traditional practice may in fact be harmful, we chose to systematically assess the available evidence to provide guidance for practice and research.